404 CHEMICAL ENGINEERING
previously cited. The great mass in the crystallizer cools slowly, the sugar separating and building up on the already formed crystals. From 8 to 10 days are required to cool completely, and frequently the cooling has to be assisted by cooling water, especially in tropical countries. The cool massecuite is pumped by large open cylinder "magma pumps" to mixers which deliver it by gravity to the centrifugal machines.
The first molasses is boiled to grain as above, and the second massecuite is put through the crystallization process described. The long cooling with constant stirring carries the separation of the sugar as far in two operations as the older process in three, the second molasses being of 25 to 28 per cent "purity." Moreover, another improvement does away with the melting and recrystallization of the second and third sugars. The second sugar, from which all possible molasses has been expelled, is placed in a mixer with fresh first molasses and after thorough mixing is again centrifuged. The expelled wash molasses goes direct to second vacuum pan. The molasses free crystals are "washed up " with water to a test of 94 to 95 per cent and go on the market as standard raw sugar. The wash goes to the first pan for first sugar. Only part of the first molasses is needed for mixing with the second sugar, the excess going to second vacuum pan. The many advantages of the modern process are evident.
A curious analogy to the boiling to "string proof " of third molasses is found in the manufacture of chrome alum. It is well known that if molasses be boiled to a high temperature and allowed to cool "taffy" results, a vifcreous noncrystalline mass. In like manner if in concentrating a solution of chrome alum, the evaporation passes a certain point, the mass on cooling assumes a glue-like consistency and refuses to crystallize. In such a case it is necessary to redissolve and reconcentrate carefully.
The manufacture of refined sugar from raw sugar is carried out in a manner very similar to the former practice for raw sugar, since quality is absolutely essential. It differs from that described in that several successive crops of crystals are taken from the liquor and that frequently, when the liquor is too impure to yield further crops of white sugar, several grades of "soft sugars" are made. When "soft sugars" are not made the impure liquor is grained at a higher temperature and the second and third sugars are remelted and returned to process for white sugar. The final molasses from raw-sugar manufacture is fit only for cattle feed or for alcohol. The final molasses from the refining process is known as "refiner's sirup " and is well known as the flavoring ingredient of the commercial so-called "corn sirups."
Deep-tank Crystallization.—The manufacture of "rock candy" from refined sugar is of interest as presenting a survival of ancient practice of crystallization formerly used for chromates and bichromates and still in use in some plants producing ferrocyanides. The concentrated liquor is run into rather deep tanks in which are suspended a large number of strings. As the liquor cools the crystals form on the strings and when crystallization is complete, usually in 3 to 5 days, the mother liquor is run off and the crystals stripped from the strings in the case of large commercial production, while for a strictly quality production such as "rock candy" threads are used, which are allowed to remain in the crystals.
This process if carried out with solutions of moderate concentration, cooled slowly, produces crystals of maximum size and beauty. The strings, however, are a constant source of trouble from tangling and from breaking during stripping. For this reason in some plants wooden strips \Y± by % in. by 4 ft. long have been substituted with good results. The crystals form as readily as on the string and are readily removed by pulling the sticks through a slot in a piece of flat iron bolted to the inside of a box for collecting the crystals.
Precipitation and Decomposition Preparatory to Crystallization.—The manufacture of sugar from sugar beets is a more complicated problem than its manu-